Solbjerg Plads 3
DK-2000 Frederiksberg
Head of Group: Hans Kurt Kvis
Telefon: +45 3815 3515
Fax: +45 3815 2500
E-mail: ma.mes@cbs.dk

Statistics
Basically, statistics deals with inference in the face of uncertainty. Thus, the ultimate goal is to contribute to decision making based on data that involves uncertainty, that is, exhibits some degree of random variation.

One important aspect of the subject, derived from this, is the study of efficient collection of data to be used in decision making.

From this rather broad definition of the subject it follows that statistics is not tied up to any particular area of application. When applied to problems within the subject area of management of business one may, schematically, conceive of management as a planning, directing, coordinating and controlling of interacting processes with the aim of obtaining more or less well formulated goals. These, in general terms, can be thought of as the "Quality of the Product".
A standard procedure, then, is to formulate a "model" for the processes. This model comprises two parts: A deterministic description of the processes via a set of fixed but unknown parameters, and a stochastic part. This enables one to view the data as the output, contaminated by random errors, from a deterministic process. By means of various principles and theories, developed and studied, in theoretical statistics, the statistician then tries to draw inferences on the parameters governing the process.

Classical applications to the management of business that the department has aimed at include: Quality control, financial planning and investment, analysis of accounting data, including auditing, and forecasting. Most of the staff works in these and related areas.

One project right now being: "Statistical Modelling in Marketing". The aim here is to develop and apply statistical methods to study the effects of advertising on individual consumers. Recent methods in data collection (eg. panel data/single-source data) have made available very comprehensive sets of data that may allow for a deeper understanding of advertising effects on the individual level than has heretofore been possible.

This project is a joint venture with Forum for Advertising Research, at the Department of Marketing.

Less classical areas taken up in recent years focus more on dynamic aspects of the processes. These areas are more closely linked to Management Science/Operations Management.

In Business Dynamics the processes are modelled by (ordinary/stochastic) differential equations. One project within this area is: "System Dynamics in Action". The object is to develop ("model") and apply methods to better ("optimally") govern the processes affecting buying and selling and transporting goods in the shipping market. This project is a joint venture with a larger private firm.

Also, within this domain the project: "Improving Supplier Relations using inventory Modelling" focuses on means to improve upon the functioning of a supply chain. In a supply chain uncertainties in e.g. arrival times of orders, size of ordered lots and quality of received lots pose a problem of optimal planning for suppliers and customers. Drawing on ideas from Supply Chain Management, Logistics, Inventory Theory, and Production Planning, the project studies this planning problem using probabilistic modelling, simulation and (statistical) analyses of available, empirical data.